•  
  •  
 

Abstract

In this paper we analyze sales data from 2016-2018 consisting of 5,547,066 records and 30 categorical variables for XYZ Furniture company (XYZ) and answer two questions of interest. The first is given the sales order data, can we predict if the shipment will miss the estimated delivery date that is generated from XYZ's internal shipment delivery date estimator. We run the data through Support Gradient Descent Boost, Random Forest, Shallow Neural Net, and Deep Neural Net models. The Deep Neural Net model provides the highest accuracy at 84.74% of predicting if a given record will miss the estimated ship date. This is an improvement of 17.97% from the current system performance. The second question of interest is when an estimated ship date is missed, can we determine the root causes of that missed shipment. We create clusters on the sales data and perform Association Rule Mining on the clusters. The rules point to Ship Plant 1, grouping deliveries, and selecting XYZ Company's transportation as the shipping method respectively as contributing factors for missed ship dates. Using the DNN solution, XYZ can expect a 17.97% increase in their prediction of missed ship dates. Investigating Ship Plant 1, grouping deliveries process, and using XYZ shipping services, XYZ may uncover some operational improvement opportunities.

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Share

COinS